remove dead code

ABL-Lab · Apr 21, 2023 · f169774 · f169774
1 parent 11c1f72
commit f169774
Showing 1 changed file with 0 additions and 87 deletions.
diff --git a/ambiguous/models/cvae.py b/ambiguous/models/cvae.py
@@ -357,90 +357,3 @@ def loss_function(self,rec,x,mu,logvar):
         rec_loss = F.mse_loss(rec, x, reduction='sum')
         loss = rec_loss + kld_loss
         return {'loss': loss, 'rec':rec_loss.detach(), 'kld':kld_loss.detach()}
-
-####################################
-class EMNIST_Encoder(nn.Module):
-    def __init__(self, latent_dim, layer_sizes, n_classes, conditional=False):
-        super(EMNIST_Encoder, self).__init__()
-        self.MLP = nn.Sequential()
-
-        self.conditional = conditional
-        if self.conditional:
-            layer_sizes[0] += n_classes
-
-        for i, (in_size, out_size) in enumerate(zip(layer_sizes[:-1], layer_sizes[1:])):
-            self.MLP.add_module(
-                name="L{:d}".format(i), module=nn.Linear(in_size, out_size))
-            self.MLP.add_module(name="A{:d}".format(i), module=nn.ReLU())
-        self.fc_mu = nn.Linear(layer_sizes[-1], latent_dim)
-        self.fc_logvar = nn.Linear(layer_sizes[-1], latent_dim)
-
-        self.device = device
-
-    def forward(self, x, c=None):
-        if self.conditional:
-            x = torch.cat([x, c], 1)       
-        h = self.MLP(x)
-        mu, logvar = self.fc_mu(h), self.fc_logvar(h)
-        return mu, logvar
-
-    def sample(self, mu, logvar):
-        std = logvar.mul(0.5).exp_()
-        if self.device == 'cuda':
-            eps = torch.cuda.FloatTensor(std.size()).normal_()
-        else:
-            eps = torch.FloatTensor(std.size()).normal_()
-        return eps.mul(std).add_(mu)
-
-class EMNIST_Decoder(nn.Module):
-    def __init__(self, latent_dim, layer_sizes, n_classes, conditional=False):
-        super(EMNIST_Decoder, self).__init__()
-
-        self.MLP = nn.Sequential()
-        self.conditional = conditional
-        if self.conditional:
-            input_size = latent_dim + n_classes
-        else:
-            input_size = latent_dim
-
-        for i, (in_size, out_size) in enumerate(zip([input_size]+layer_sizes[:-1], layer_sizes)):
-            self.MLP.add_module(
-                name="L{:d}".format(i), module=nn.Linear(in_size, out_size))
-            if i+1 < len(layer_sizes):
-                self.MLP.add_module(name="A{:d}".format(i), module=nn.ReLU())
-            else:
-                self.MLP.add_module(name="sigmoid", module=nn.Sigmoid())
-    def forward(self, z, c):
-        if self.conditional:
-            z = torch.cat([z, c], 1)
-        rec = self.MLP(z)
-        return rec   
-
-
-class EMNIST_EncoderV1(nn.Module):
-    def __init__(self, latent_dim, layer_sizes):
-        super(EMNIST_EncoderV1, self).__init__()
-        self.MLP = nn.Sequential()
-        for i, (in_size, out_size) in enumerate(zip(layer_sizes[:-1], layer_sizes[1:])):
-            self.MLP.add_module(
-                name="L{:d}".format(i), module=nn.Linear(in_size, out_size))
-            self.MLP.add_module(name="A{:d}".format(i), module=nn.ReLU())
-        self.fc_mu = nn.Linear(layer_sizes[-1], latent_dim)
-        self.fc_logvar = nn.Linear(layer_sizes[-1], latent_dim)
-
-        self.device = device
-
-    def forward(self, x):     
-        h = self.MLP(x)
-        mu, logvar = self.fc_mu(h), self.fc_logvar(h)
-        return mu, logvar
-
-    def sample(self, mu, logvar):
-        std = logvar.mul(0.5).exp_()
-        if self.device == 'cuda':
-            eps = torch.cuda.FloatTensor(std.size()).normal_()
-        else:
-            eps = torch.FloatTensor(std.size()).normal_()
-        return eps.mul(std).add_(mu)
-
-